#ifndef LLVM_ANALYSIS_TARGETTRANSFORMINFO_H
#define LLVM_ANALYSIS_TARGETTRANSFORMINFO_H
-#include "llvm/IR/Intrinsics.h"
+#include "llvm/ADT/Optional.h"
#include "llvm/IR/IntrinsicInst.h"
+#include "llvm/IR/Intrinsics.h"
#include "llvm/Pass.h"
#include "llvm/Support/DataTypes.h"
+#include <functional>
namespace llvm {
class Function;
class GlobalValue;
class Loop;
+class PreservedAnalyses;
class Type;
class User;
class Value;
/// implementaion that encodes appropriate costs for their target.
template <typename T> TargetTransformInfo(T Impl);
+ /// \brief Construct a baseline TTI object using a minimal implementation of
+ /// the \c Concept API below.
+ ///
+ /// The TTI implementation will reflect the information in the DataLayout
+ /// provided if non-null.
+ explicit TargetTransformInfo(const DataLayout &DL);
+
// Provide move semantics.
TargetTransformInfo(TargetTransformInfo &&Arg);
TargetTransformInfo &operator=(TargetTransformInfo &&RHS);
// out-of-line.
~TargetTransformInfo();
+ /// \brief Handle the invalidation of this information.
+ ///
+ /// When used as a result of \c TargetIRAnalysis this method will be called
+ /// when the function this was computed for changes. When it returns false,
+ /// the information is preserved across those changes.
+ bool invalidate(Function &, const PreservedAnalyses &) {
+ // FIXME: We should probably in some way ensure that the subtarget
+ // information for a function hasn't changed.
+ return false;
+ }
+
/// \name Generic Target Information
/// @{
/// comments for a detailed explanation of the cost values.
unsigned getUserCost(const User *U) const;
- /// \brief hasBranchDivergence - Return true if branch divergence exists.
+ /// \brief Return true if branch divergence exists.
+ ///
/// Branch divergence has a significantly negative impact on GPU performance
/// when threads in the same wavefront take different paths due to conditional
/// branches.
bool hasBranchDivergence() const;
+ /// \brief Returns whether V is a source of divergence.
+ ///
+ /// This function provides the target-dependent information for
+ /// the target-independent DivergenceAnalysis. DivergenceAnalysis first
+ /// builds the dependency graph, and then runs the reachability algorithm
+ /// starting with the sources of divergence.
+ bool isSourceOfDivergence(const Value *V) const;
+
/// \brief Test whether calls to a function lower to actual program function
/// calls.
///
/// Parameters that control the generic loop unrolling transformation.
struct UnrollingPreferences {
- /// The cost threshold for the unrolled loop, compared to
- /// CodeMetrics.NumInsts aggregated over all basic blocks in the loop body.
- /// The unrolling factor is set such that the unrolled loop body does not
- /// exceed this cost. Set this to UINT_MAX to disable the loop body cost
+ /// The cost threshold for the unrolled loop. Should be relative to the
+ /// getUserCost values returned by this API, and the expectation is that
+ /// the unrolled loop's instructions when run through that interface should
+ /// not exceed this cost. However, this is only an estimate. Also, specific
+ /// loops may be unrolled even with a cost above this threshold if deemed
+ /// profitable. Set this to UINT_MAX to disable the loop body cost
/// restriction.
unsigned Threshold;
+ /// If complete unrolling will reduce the cost of the loop below its
+ /// expected dynamic cost while rolled by this percentage, apply a discount
+ /// (below) to its unrolled cost.
+ unsigned PercentDynamicCostSavedThreshold;
+ /// The discount applied to the unrolled cost when the *dynamic* cost
+ /// savings of unrolling exceed the \c PercentDynamicCostSavedThreshold.
+ unsigned DynamicCostSavingsDiscount;
/// The cost threshold for the unrolled loop when optimizing for size (set
/// to UINT_MAX to disable).
unsigned OptSizeThreshold;
/// loop body even when the number of loop iterations is not known at
/// compile time).
bool Runtime;
+ /// Allow emitting expensive instructions (such as divisions) when computing
+ /// the trip count of a loop for runtime unrolling.
+ bool AllowExpensiveTripCount;
};
/// \brief Get target-customized preferences for the generic loop unrolling
/// transformation. The caller will initialize UP with the current
/// target-independent defaults.
- void getUnrollingPreferences(const Function *F, Loop *L,
- UnrollingPreferences &UP) const;
+ void getUnrollingPreferences(Loop *L, UnrollingPreferences &UP) const;
/// @}
/// mode is legal for a load/store of any legal type.
/// TODO: Handle pre/postinc as well.
bool isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV, int64_t BaseOffset,
- bool HasBaseReg, int64_t Scale) const;
+ bool HasBaseReg, int64_t Scale,
+ unsigned AddrSpace = 0) const;
/// \brief Return true if the target works with masked instruction
/// AVX2 allows masks for consecutive load and store for i32 and i64 elements.
/// If the AM is not supported, it returns a negative value.
/// TODO: Handle pre/postinc as well.
int getScalingFactorCost(Type *Ty, GlobalValue *BaseGV, int64_t BaseOffset,
- bool HasBaseReg, int64_t Scale) const;
+ bool HasBaseReg, int64_t Scale,
+ unsigned AddrSpace = 0) const;
/// \brief Return true if it's free to truncate a value of type Ty1 to type
/// Ty2. e.g. On x86 it's free to truncate a i32 value in register EAX to i16
/// by referencing its sub-register AX.
bool isTruncateFree(Type *Ty1, Type *Ty2) const;
+ /// \brief Return true if it is profitable to hoist instruction in the
+ /// then/else to before if.
+ bool isProfitableToHoist(Instruction *I) const;
+
/// \brief Return true if this type is legal.
bool isTypeLegal(Type *Ty) const;
/// target.
bool shouldBuildLookupTables() const;
+ /// \brief Don't restrict interleaved unrolling to small loops.
+ bool enableAggressiveInterleaving(bool LoopHasReductions) const;
+
/// \brief Return hardware support for population count.
PopcntSupportKind getPopcntSupport(unsigned IntTyWidthInBit) const;
/// \brief Return true if the hardware has a fast square-root instruction.
bool haveFastSqrt(Type *Ty) const;
+ /// \brief Return the expected cost of supporting the floating point operation
+ /// of the specified type.
+ unsigned getFPOpCost(Type *Ty) const;
+
/// \brief Return the expected cost of materializing for the given integer
/// immediate of the specified type.
unsigned getIntImmCost(const APInt &Imm, Type *Ty) const;
/// \return The maximum interleave factor that any transform should try to
/// perform for this target. This number depends on the level of parallelism
/// and the number of execution units in the CPU.
- unsigned getMaxInterleaveFactor() const;
+ unsigned getMaxInterleaveFactor(unsigned VF) const;
/// \return The expected cost of arithmetic ops, such as mul, xor, fsub, etc.
unsigned
unsigned getMaskedMemoryOpCost(unsigned Opcode, Type *Src, unsigned Alignment,
unsigned AddressSpace) const;
+ /// \return The cost of the interleaved memory operation.
+ /// \p Opcode is the memory operation code
+ /// \p VecTy is the vector type of the interleaved access.
+ /// \p Factor is the interleave factor
+ /// \p Indices is the indices for interleaved load members (as interleaved
+ /// load allows gaps)
+ /// \p Alignment is the alignment of the memory operation
+ /// \p AddressSpace is address space of the pointer.
+ unsigned getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
+ unsigned Factor,
+ ArrayRef<unsigned> Indices,
+ unsigned Alignment,
+ unsigned AddressSpace) const;
+
/// \brief Calculate the cost of performing a vector reduction.
///
/// This is the cost of reducing the vector value of type \p Ty to a scalar
unsigned getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
ArrayRef<Type *> Tys) const;
+ /// \returns The cost of Call instructions.
+ unsigned getCallInstrCost(Function *F, Type *RetTy,
+ ArrayRef<Type *> Tys) const;
+
/// \returns The number of pieces into which the provided type must be
/// split during legalization. Zero is returned when the answer is unknown.
unsigned getNumberOfParts(Type *Tp) const;
Value *getOrCreateResultFromMemIntrinsic(IntrinsicInst *Inst,
Type *ExpectedType) const;
+ /// \returns True if the two functions have compatible attributes for inlining
+ /// purposes.
+ bool hasCompatibleFunctionAttributes(const Function *Caller,
+ const Function *Callee) const;
+
/// @}
private:
class TargetTransformInfo::Concept {
public:
virtual ~Concept() = 0;
-
+ virtual const DataLayout &getDataLayout() const = 0;
virtual unsigned getOperationCost(unsigned Opcode, Type *Ty, Type *OpTy) = 0;
virtual unsigned getGEPCost(const Value *Ptr,
ArrayRef<const Value *> Operands) = 0;
ArrayRef<const Value *> Arguments) = 0;
virtual unsigned getUserCost(const User *U) = 0;
virtual bool hasBranchDivergence() = 0;
+ virtual bool isSourceOfDivergence(const Value *V) = 0;
virtual bool isLoweredToCall(const Function *F) = 0;
- virtual void getUnrollingPreferences(const Function *F, Loop *L,
- UnrollingPreferences &UP) = 0;
+ virtual void getUnrollingPreferences(Loop *L, UnrollingPreferences &UP) = 0;
virtual bool isLegalAddImmediate(int64_t Imm) = 0;
virtual bool isLegalICmpImmediate(int64_t Imm) = 0;
virtual bool isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV,
int64_t BaseOffset, bool HasBaseReg,
- int64_t Scale) = 0;
+ int64_t Scale,
+ unsigned AddrSpace) = 0;
virtual bool isLegalMaskedStore(Type *DataType, int Consecutive) = 0;
virtual bool isLegalMaskedLoad(Type *DataType, int Consecutive) = 0;
virtual int getScalingFactorCost(Type *Ty, GlobalValue *BaseGV,
int64_t BaseOffset, bool HasBaseReg,
- int64_t Scale) = 0;
+ int64_t Scale, unsigned AddrSpace) = 0;
virtual bool isTruncateFree(Type *Ty1, Type *Ty2) = 0;
+ virtual bool isProfitableToHoist(Instruction *I) = 0;
virtual bool isTypeLegal(Type *Ty) = 0;
virtual unsigned getJumpBufAlignment() = 0;
virtual unsigned getJumpBufSize() = 0;
virtual bool shouldBuildLookupTables() = 0;
+ virtual bool enableAggressiveInterleaving(bool LoopHasReductions) = 0;
virtual PopcntSupportKind getPopcntSupport(unsigned IntTyWidthInBit) = 0;
virtual bool haveFastSqrt(Type *Ty) = 0;
+ virtual unsigned getFPOpCost(Type *Ty) = 0;
virtual unsigned getIntImmCost(const APInt &Imm, Type *Ty) = 0;
virtual unsigned getIntImmCost(unsigned Opc, unsigned Idx, const APInt &Imm,
Type *Ty) = 0;
const APInt &Imm, Type *Ty) = 0;
virtual unsigned getNumberOfRegisters(bool Vector) = 0;
virtual unsigned getRegisterBitWidth(bool Vector) = 0;
- virtual unsigned getMaxInterleaveFactor() = 0;
+ virtual unsigned getMaxInterleaveFactor(unsigned VF) = 0;
virtual unsigned
getArithmeticInstrCost(unsigned Opcode, Type *Ty, OperandValueKind Opd1Info,
OperandValueKind Opd2Info,
virtual unsigned getMaskedMemoryOpCost(unsigned Opcode, Type *Src,
unsigned Alignment,
unsigned AddressSpace) = 0;
+ virtual unsigned getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
+ unsigned Factor,
+ ArrayRef<unsigned> Indices,
+ unsigned Alignment,
+ unsigned AddressSpace) = 0;
virtual unsigned getReductionCost(unsigned Opcode, Type *Ty,
bool IsPairwiseForm) = 0;
virtual unsigned getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
ArrayRef<Type *> Tys) = 0;
+ virtual unsigned getCallInstrCost(Function *F, Type *RetTy,
+ ArrayRef<Type *> Tys) = 0;
virtual unsigned getNumberOfParts(Type *Tp) = 0;
virtual unsigned getAddressComputationCost(Type *Ty, bool IsComplex) = 0;
virtual unsigned getCostOfKeepingLiveOverCall(ArrayRef<Type *> Tys) = 0;
MemIntrinsicInfo &Info) = 0;
virtual Value *getOrCreateResultFromMemIntrinsic(IntrinsicInst *Inst,
Type *ExpectedType) = 0;
+ virtual bool hasCompatibleFunctionAttributes(const Function *Caller,
+ const Function *Callee) const = 0;
};
template <typename T>
Model(T Impl) : Impl(std::move(Impl)) {}
~Model() override {}
+ const DataLayout &getDataLayout() const override {
+ return Impl.getDataLayout();
+ }
+
unsigned getOperationCost(unsigned Opcode, Type *Ty, Type *OpTy) override {
return Impl.getOperationCost(Opcode, Ty, OpTy);
}
}
unsigned getUserCost(const User *U) override { return Impl.getUserCost(U); }
bool hasBranchDivergence() override { return Impl.hasBranchDivergence(); }
+ bool isSourceOfDivergence(const Value *V) override {
+ return Impl.isSourceOfDivergence(V);
+ }
bool isLoweredToCall(const Function *F) override {
return Impl.isLoweredToCall(F);
}
- void getUnrollingPreferences(const Function *F, Loop *L,
- UnrollingPreferences &UP) override {
- return Impl.getUnrollingPreferences(F, L, UP);
+ void getUnrollingPreferences(Loop *L, UnrollingPreferences &UP) override {
+ return Impl.getUnrollingPreferences(L, UP);
}
bool isLegalAddImmediate(int64_t Imm) override {
return Impl.isLegalAddImmediate(Imm);
return Impl.isLegalICmpImmediate(Imm);
}
bool isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV, int64_t BaseOffset,
- bool HasBaseReg, int64_t Scale) override {
+ bool HasBaseReg, int64_t Scale,
+ unsigned AddrSpace) override {
return Impl.isLegalAddressingMode(Ty, BaseGV, BaseOffset, HasBaseReg,
- Scale);
+ Scale, AddrSpace);
}
bool isLegalMaskedStore(Type *DataType, int Consecutive) override {
return Impl.isLegalMaskedStore(DataType, Consecutive);
return Impl.isLegalMaskedLoad(DataType, Consecutive);
}
int getScalingFactorCost(Type *Ty, GlobalValue *BaseGV, int64_t BaseOffset,
- bool HasBaseReg, int64_t Scale) override {
- return Impl.getScalingFactorCost(Ty, BaseGV, BaseOffset, HasBaseReg, Scale);
+ bool HasBaseReg, int64_t Scale,
+ unsigned AddrSpace) override {
+ return Impl.getScalingFactorCost(Ty, BaseGV, BaseOffset, HasBaseReg,
+ Scale, AddrSpace);
}
bool isTruncateFree(Type *Ty1, Type *Ty2) override {
return Impl.isTruncateFree(Ty1, Ty2);
}
+ bool isProfitableToHoist(Instruction *I) override {
+ return Impl.isProfitableToHoist(I);
+ }
bool isTypeLegal(Type *Ty) override { return Impl.isTypeLegal(Ty); }
unsigned getJumpBufAlignment() override { return Impl.getJumpBufAlignment(); }
unsigned getJumpBufSize() override { return Impl.getJumpBufSize(); }
bool shouldBuildLookupTables() override {
return Impl.shouldBuildLookupTables();
}
+ bool enableAggressiveInterleaving(bool LoopHasReductions) override {
+ return Impl.enableAggressiveInterleaving(LoopHasReductions);
+ }
PopcntSupportKind getPopcntSupport(unsigned IntTyWidthInBit) override {
return Impl.getPopcntSupport(IntTyWidthInBit);
}
bool haveFastSqrt(Type *Ty) override { return Impl.haveFastSqrt(Ty); }
+
+ unsigned getFPOpCost(Type *Ty) override {
+ return Impl.getFPOpCost(Ty);
+ }
+
unsigned getIntImmCost(const APInt &Imm, Type *Ty) override {
return Impl.getIntImmCost(Imm, Ty);
}
unsigned getRegisterBitWidth(bool Vector) override {
return Impl.getRegisterBitWidth(Vector);
}
- unsigned getMaxInterleaveFactor() override {
- return Impl.getMaxInterleaveFactor();
+ unsigned getMaxInterleaveFactor(unsigned VF) override {
+ return Impl.getMaxInterleaveFactor(VF);
}
unsigned
getArithmeticInstrCost(unsigned Opcode, Type *Ty, OperandValueKind Opd1Info,
unsigned AddressSpace) override {
return Impl.getMaskedMemoryOpCost(Opcode, Src, Alignment, AddressSpace);
}
+ unsigned getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
+ unsigned Factor,
+ ArrayRef<unsigned> Indices,
+ unsigned Alignment,
+ unsigned AddressSpace) override {
+ return Impl.getInterleavedMemoryOpCost(Opcode, VecTy, Factor, Indices,
+ Alignment, AddressSpace);
+ }
unsigned getReductionCost(unsigned Opcode, Type *Ty,
bool IsPairwiseForm) override {
return Impl.getReductionCost(Opcode, Ty, IsPairwiseForm);
ArrayRef<Type *> Tys) override {
return Impl.getIntrinsicInstrCost(ID, RetTy, Tys);
}
+ unsigned getCallInstrCost(Function *F, Type *RetTy,
+ ArrayRef<Type *> Tys) override {
+ return Impl.getCallInstrCost(F, RetTy, Tys);
+ }
unsigned getNumberOfParts(Type *Tp) override {
return Impl.getNumberOfParts(Tp);
}
Type *ExpectedType) override {
return Impl.getOrCreateResultFromMemIntrinsic(Inst, ExpectedType);
}
+ bool hasCompatibleFunctionAttributes(const Function *Caller,
+ const Function *Callee) const override {
+ return Impl.hasCompatibleFunctionAttributes(Caller, Callee);
+ }
};
template <typename T>
TargetTransformInfo::TargetTransformInfo(T Impl)
: TTIImpl(new Model<T>(Impl)) {}
+/// \brief Analysis pass providing the \c TargetTransformInfo.
+///
+/// The core idea of the TargetIRAnalysis is to expose an interface through
+/// which LLVM targets can analyze and provide information about the middle
+/// end's target-independent IR. This supports use cases such as target-aware
+/// cost modeling of IR constructs.
+///
+/// This is a function analysis because much of the cost modeling for targets
+/// is done in a subtarget specific way and LLVM supports compiling different
+/// functions targeting different subtargets in order to support runtime
+/// dispatch according to the observed subtarget.
+class TargetIRAnalysis {
+public:
+ typedef TargetTransformInfo Result;
+
+ /// \brief Opaque, unique identifier for this analysis pass.
+ static void *ID() { return (void *)&PassID; }
+
+ /// \brief Provide access to a name for this pass for debugging purposes.
+ static StringRef name() { return "TargetIRAnalysis"; }
+
+ /// \brief Default construct a target IR analysis.
+ ///
+ /// This will use the module's datalayout to construct a baseline
+ /// conservative TTI result.
+ TargetIRAnalysis();
+
+ /// \brief Construct an IR analysis pass around a target-provide callback.
+ ///
+ /// The callback will be called with a particular function for which the TTI
+ /// is needed and must return a TTI object for that function.
+ TargetIRAnalysis(std::function<Result(Function &)> TTICallback);
+
+ // Value semantics. We spell out the constructors for MSVC.
+ TargetIRAnalysis(const TargetIRAnalysis &Arg)
+ : TTICallback(Arg.TTICallback) {}
+ TargetIRAnalysis(TargetIRAnalysis &&Arg)
+ : TTICallback(std::move(Arg.TTICallback)) {}
+ TargetIRAnalysis &operator=(const TargetIRAnalysis &RHS) {
+ TTICallback = RHS.TTICallback;
+ return *this;
+ }
+ TargetIRAnalysis &operator=(TargetIRAnalysis &&RHS) {
+ TTICallback = std::move(RHS.TTICallback);
+ return *this;
+ }
+
+ Result run(Function &F);
+
+private:
+ static char PassID;
+
+ /// \brief The callback used to produce a result.
+ ///
+ /// We use a completely opaque callback so that targets can provide whatever
+ /// mechanism they desire for constructing the TTI for a given function.
+ ///
+ /// FIXME: Should we really use std::function? It's relatively inefficient.
+ /// It might be possible to arrange for even stateful callbacks to outlive
+ /// the analysis and thus use a function_ref which would be lighter weight.
+ /// This may also be less error prone as the callback is likely to reference
+ /// the external TargetMachine, and that reference needs to never dangle.
+ std::function<Result(Function &)> TTICallback;
+
+ /// \brief Helper function used as the callback in the default constructor.
+ static Result getDefaultTTI(Function &F);
+};
+
/// \brief Wrapper pass for TargetTransformInfo.
///
/// This pass can be constructed from a TTI object which it stores internally
/// and is queried by passes.
class TargetTransformInfoWrapperPass : public ImmutablePass {
- TargetTransformInfo TTI;
+ TargetIRAnalysis TIRA;
+ Optional<TargetTransformInfo> TTI;
virtual void anchor();
/// Use the constructor below or call one of the creation routines.
TargetTransformInfoWrapperPass();
- explicit TargetTransformInfoWrapperPass(TargetTransformInfo TTI);
+ explicit TargetTransformInfoWrapperPass(TargetIRAnalysis TIRA);
- TargetTransformInfo &getTTI() { return TTI; }
- const TargetTransformInfo &getTTI() const { return TTI; }
+ TargetTransformInfo &getTTI(Function &F);
};
-/// \brief Create the base case instance of a pass in the TTI analysis group.
+/// \brief Create an analysis pass wrapper around a TTI object.
///
-/// This class provides the base case for the stack of TTI analyzes. It doesn't
-/// delegate to anything and uses the STTI and VTTI objects passed in to
-/// satisfy the queries.
-ImmutablePass *createNoTargetTransformInfoPass(const DataLayout *DL);
+/// This analysis pass just holds the TTI instance and makes it available to
+/// clients.
+ImmutablePass *createTargetTransformInfoWrapperPass(TargetIRAnalysis TIRA);
} // End llvm namespace
projects / oota-llvm.git / blobdiff